research article neutrophil gelatinase-associated lipocalin in...

Hindawi Publishing CorporationJournal of TransplantationVolume 2013, Article ID 650123, 12 pageshttp://dx.doi.org/10.1155/2013/650123

Research ArticleNeutrophil Gelatinase-Associated Lipocalin in KidneyTransplantation Is an Early Marker of Graft Dysfunction and IsAssociated with One-Year Renal Function

Isabel Fonseca,1,2 José Carlos Oliveira,3 Manuela Almeida,1,2 Madalena Cruz,3

Anabela Malho,1 La Salete Martins,1,2 Leonídio Dias,1 Sofia Pedroso,1

Josefina Santos,1,2 Luísa Lobato,1,2 António Castro Henriques,1,2 and Denisa Mendonça4,5

1 Department of Nephrology and Kidney Transplantation, Centro Hospitalar do Porto, Hospital de Santo Antonio,4099-001 Porto, Portugal

2 Unit for Multidisciplinary Investigation in Biomedicine (UMIB), 4099-003 Porto, Portugal3 Department of Chemical Chemistry, Centro Hospitalar do Porto, Hospital de Santo Antonio, 4099-001 Porto, Portugal4 Institute of Public Health (ISPUP), University of Porto, 4050-600 Porto, Portugal5 Department of Population Studies, Institute of Biomedical Sciences Abel Salazar (ICBAS), University of Porto,4050-313 Porto, Portugal

Correspondence should be addressed to Isabel Fonseca; [email protected]

Received 30 July 2013; Revised 7 September 2013; Accepted 20 September 2013

Academic Editor: Bruce Kaplan

Copyright © 2013 Isabel Fonseca et al. This is an open access article distributed under the Creative Commons Attribution License,which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Urinary neutrophil gelatinase-associated lipocalin (uNGAL) has been suggested as potential early marker of delayed graft function(DGF) following kidney transplantation (KTx). We conducted a prospective study in 40 consecutive KTx recipients to evaluateserial changes of uNGALwithin the first week after KTx and assess its performance in predicting DGF (dialysis requirement duringinitial posttransplant week) and graft function throughout first year. Urine samples were collected on post-KTx days 0, 1, 2, 4, and7. Linear mixed and multivariable regression models, receiver-operating characteristic (ROC), and areas under ROC curves wereused. At all-time points, mean uNGAL levels were significantly higher in patients developing DGF (𝑛 = 18). Shortly after KTx(3–6 h), uNGAL values were higher in DGF recipients (on average +242 ng/mL, considering mean dialysis time of 4.1 years) androse further in following days, contrasting with prompt function recipients. Day-1 uNGAL levels accurately predicted DGF (AUC-ROC = 0.93), with a performance higher than serum creatinine (AUC-ROC = 0.76), and similar to cystatin C (AUC-ROC = 0.95).Multivariable analyses revealed that uNGAL levels at days 4 and 7 were strongly associated with one-year serum creatinine. UrinaryNGAL is an early marker of graft injury and is independently associated with dialysis requirement within one week after KTx andone-year graft function.

1. Introduction

Delayed graft function (DGF) is an important complicationof kidney transplantation (KTx) that adversely affects allo-graft survival. Despite substantial improvements in the fieldof KTx, the incidence of DGF is rising with the growingpractice of accepting expanded criteria donors to increasetransplantation rates [1–6]. Delayed graft function predis-poses kidney graft to acute and chronic rejection, contributesto progressive allograft dysfunction, and increases the risk ofpremature graft loss [7–11].

Reliable biomarkers enabling early discrimination ofDGF in KTx are lacking, which impairs timely therapeuticinterventions. Traditionally, acute graft dysfunction is diag-nosed by measuring serum creatinine, but this parameteris an unreliable indicator of kidney function during anepisode of acute injury [12]. One of the most promisingbiomarkers of acute kidney injury is neutrophil gelatinase-associated lipocalin (NGAL), which is released to bloodfrom activated neutrophils during inflammatory processes.In steady situations, this lipocalin is found in urine only intrace. Massive NGAL quantities excreted in urine (uNGAL)

2 Journal of Transplantation

usually indicate damage of proximal tubular cells [13–15].Graft injury due to ischemia-reperfusion is an inevitableconsequence of KTx procedure and can result in varyingdegrees of early graft dysfunction, which can be considereda form of posttransplantation acute kidney injury. For thisreason, several studies investigated the utility of NGAL forthe diagnostic and prognostic of acute graft dysfunctionfollowing KTx [16–27]. Recently, the prognostic value ofuNGAL on graft function at one-year after transplantationwas also examined and presented conflicting findings [22,28].

In order to support the usefulness of uNGAL as a reliablemarker of graft injury and to clarify the role of this promisingbiomarker in the prediction of kidney function beyond thefirst week after transplant, we conducted a prospective studyto

(a) evaluate longitudinal changes of uNGAL levels overthe first week after KTx and identify factors associatedwith these changes;

(b) assess the performance of uNGAL in predicting DGF(defined as the requirement for dialysiswithin the first7 days after transplantation);

(c) appraise the long-term prognostic value of uNGALmeasured within one week posttransplantation onkidney allograft function, evaluated by one-yearserum creatinine.

2. Subjects and Methods

Consecutive patients with end-stage renal disease, undergo-ing living or deceased KTx at Centro Hospitalar do Porto,fromDecember 2010 toMay 2011 were prospectively enrolled.Recruitment excluded patients less than 18 years old andthose who required a combined pancreas or liver KTx. Aftertransplant, patients with primary graft failure related tosurgical causes were excluded. This study was approved byInstitutional Review Board of Centro Hospitalar do Porto.Each participant provided written informed consent beforeenrolment.

2.1. Sample Collection and Measurements. Urine samples forNGAL determination were collected 3 to 6 h after surgery(uNGAL0 or baseline); on the following morning, nearly8 to 12 h after graft reperfusion (uNGAL1 or first day);and then at second (uNGAL2), fourth (uNGAL4), andseventh days (uNGAL7), for a total of five samples for eachpatient. The same laboratory technician, who was blinded topatient information, performed uNGALmeasurements usinga two-step chemiluminescent microparticle immunoassayon a standardized clinical platform (ARCHITECT, AbbottDiagnostics).

Serum creatinine levels were determined preoperatively,daily until hospital discharge, and at 1, 3, 6, and 12monthsafter transplantation to evaluate later graft function. Serumcreatinine measurements were performed by Jaffe method(RocheDiagnostics). Cystatin Cwasmeasuredwith a particle

enhanced immunonephelometric method (Siemens Diag-nostics) at the same time points as uNGAL, except forbaseline.

2.2. Definitions. Delayed graft function was defined, accord-ing to United Network for Organ Sharing, as the requirementfor dialysis within the first seven days after KTx, due toan absence or irrelevant improvement in graft function.Complementarily, graft function was considered “prompt”(non-DGF) if no dialysis was required during the first weekafter transplantation.

Acute Rejection was defined as either biopsy-provenrejection or antirejection treatment without biopsy.

Estimated Glomerular Filtration Rate (eGFR) was calcu-lated using theRule’s refittedMDRD formula [29], consideredto have an improved diagnostic performance and betteraccuracy of the true GFR in KTx recipients [30].

Creatinine Reduction Rate (%) was calculated as thedifference between serum creatinine at day 2 (or day 4) andday 1, divided by serum creatinine at day 1, multiplied by 100.

Graft function at one year was evaluated by the averageof the two serum creatinine levels measured closer to oneyear after KTx (e.g., by the average values seen at 12 and13months). It was thought that this would reflect moreaccurately the usual graft function, since a single measurecould bemore easily inflated by acute situations, like a urinaryinfection for example. Two grafts were lost at seventh andeighth months and the last serum creatinine presented bythese patients prior to dialysis restart was considered as beingthe one-year creatinine.

2.3. Statistics. Kolmogorov-Smirnov test was performed toassess deviation from normal distribution. Quantitative vari-ables were summarized asmean and standard deviation (SD),or asmedian and 25th–75th quartiles (interquartile range) forvariableswith skewed distribution. Categorical variableswerereported as percentages.

Statistical analysis was performed in five steps. Firstly,a cross-sectional bivariate analysis was done to comparegroups and study the association between uNGAL anddemographic/clinical variables. Continuous variables werecompared using either parametric (t-test) or nonparamet-ric (Mann-Whitney) tests. Associations between categoricalvariables were analyzed using the 𝜒2 test and Fisher’s exacttest as appropriate. Correlations between uNGAL and con-tinuous variables were assessed using Pearson correlationand uNGAL levels were log-transformed (ln) before analysis.Spearman correlationwas used to analyze uNGAL and serumcreatinine reduction ratio on posttransplant days 2 and 4.

Secondly, we used a longitudinal analysis to study uNGALkinetics and modelling it as a response variable on time. Alinear mixed-effects model was used to study the associationof DGF with serial changes of uNGAL (log-transformed),controlling for donor status (living/deceased), recipient’s age,time on dialysis, and time measurement of uNGAL. Theinteraction between DGF and uNGAL time measurementwas included in the model, as such a significant interactionwould suggest that DGF affects the uNGAL levels trajectory.

Journal of Transplantation 3

Thirdly, receiver-operating characteristics analysis wasperformed to estimate the sensitivity and specificity ofuNGAL (as well as serum creatinine and cystatin C) topredict DGF. The optimal cut-off points were determined bymaximizing the sum of sensitivity and specificity.

Fourthly, multivariable logistic regression analysis wasundertaken to evaluate whether uNGAL levels were indepen-dently associated with DGF. Pretransplant variables knownto be associated with DGF and considered potential con-founders were included in the models. To avoid collinearityeach time point uNGAL was included separately in differentmodels. The final models were fitted using a backwardselection procedure.

Fifthly, multivariable linear regression was used todescribe the independent association of uNGAL with renalfunction at 12months evaluated by serumcreatinine, adjustedfor the variables that usually predict graft function, includingdonor status, rehospitalizations, and acute rejection episodesthroughout the first year. Linear regression models usedlog-transformed uNGAL and serum creatinine levels. As inlogistic models, uNGAL at each time point were includedseparately in models to avoid collinearity.

All statistical analyses were done with SPSS version 20.0and a significance level of 0.05 was considered.

3. Results

During time recruitment, 42 patients were enrolled. Tworecipients had renal artery thrombosis and were excluded inthe first two posttransplantation days. Therefore our studysample included 40 recipients. Baseline data are shown inTable 1.

3.1. Urinary NGAL. The first urine sample (uNGAL0) wasobtained from 30 patients. On the following days, urinesamples were collected from 35 patients at the first, second,and seventh days, and from 36 patients at the fourth day.All of our subjects provided at least two urine samples.Only, one patient provided merely two urine samples and theremaining 39 subjects provided 3 ormore urine samples (with20 patients providing all five samples).

Daily median uNGAL levels did not differ significantlybetween male and female recipients, except for the seventhday where female uNGAL levels were significantly higher.Concerning donor status, uNGAL levels were higher indeceased donor recipients at all-time points, but only sta-tistically significant at second day. Except for the seventhday, uNGAL levelswere significantly andpositively correlatedwith cold ischemia time (𝑟 = 0.45, 𝑃 = 0.02; 𝑟 = 0.36,𝑃 = 0.04; 𝑟 = 0.56, 𝑃 = 0.001; 𝑟 = 0.46, 𝑃 = 0.006, resp.,at baseline, first, second, and fourth days).

At most time points, uNGAL was positively and signif-icantly correlated with recipient age (𝑟 = 0.39, 𝑃 = 0.02;𝑟 = 0.39, 𝑃 = 0.02; 𝑟 = 0.44, 𝑃 = 0.007; resp., at first, second,and seventh days) and pretransplant dialysis time (𝑟 = 0.48,𝑃 = 0.008; 𝑟 = 0.37, 𝑃 = 0.03; 𝑟 = 0.43, 𝑃 = 0.01; 𝑟 = 0.33,𝑃 = 0.024; resp., at baseline, first, second, and seventh days).

No significant correlation was found with HLA mismatchesand with donor age and serum creatinine.

Urinary NGAL levels were significantly and positivelycorrelated with serum creatinine at all-time points (data notshown). Furthermore, except for uNGAL0, all the remaininguNGAL levels were significantly and negatively correlatedwith changes in serum creatinine between the second andfirst days, and also between the fourth and the first days: loweruNGAL values were associated with higher reductions ratesin serum creatinine (data not shown).

Median length of hospitalization after transplantationwas12 days (IQR: 7–22) and uNGAL levels were highly correlatedwith length of hospital stay at all-time points (𝑟 = 0.48, 𝑃 =0.002; 𝑟 = 0.64, 𝑃 < 0.001; 𝑟 = 0.79, 𝑃 < 0.001; 𝑟 = 0.77,𝑃 < 0.001; 𝑟 = 0.82,𝑃 < 0.001, resp., at baseline, first, second,fourth, and seventh days).

3.2. DGF and uNGAL Longitudinal Changes. Eighteen recip-ients (45%) had DGF, three of these were from living donors,and 22 (55%) had prompt graft function. Concerning tradi-tional predictors of DGF and except for cold ischemia time,no significant differences were found between DGF/non-DGF in relation to baseline characteristics and inductiontherapy (Table 1). Mean age was significantly higher inpatients with DGF (56 (11) versus 43 (16) years in non-DGFrecipients, 𝑃 = 0.006). As expected, patients with DGF hadhigher serum creatinine levels (Table 2) and lower creatininereduction ratios on posttransplant days 2 and 4.

Similar to serum creatinine, median uNGAL concen-trations were consistently higher in DGF group comparedwith non-DGF group at all measured time points (Table 2and Figure 1). In patients with prompt graft function, thelongitudinal changes of uNGAL were characterized by aninitial phase with a rapid decline and then a phase with aslower decrease continuing throughout the posttransplantweek.This pattern of changes was different inDGF recipients:uNGAL levels increased frombaseline to the followingmorn-ing after transplantation and remained elevated throughoutmost of the follow-up period.

A linear mixed-effects model was used to study theassociation of DGF with longitudinal changes of uNGAL,controlling for variables found to be associated with uNGALby bivariate analysis. Pretransplant time on dialysis, timemeasurement of uNGAL, and DGF were independentlyassociated with uNGAL levels. Adjusting for the remainingvariables, donor status and recipient age lost their statisticalsignificance andwere removed from the finalmodel (Table 3).Delayed graft function was significantly associated withuNGAL levels, with prompt function recipients having onaverage lower levels of uNGAL at all-time points. Accordingto our estimation, for a patient with dialysis time of approx-imately 4.1 years, the initial values of uNGAL (3–6 h aftertransplantation) are about 242 ng/mL higher in patients whowent on to develop DGF, and these values will rise evenmore in the following days. A significant interaction betweentime of measurement and DGF confirmed that longitudinalchanges of uNGAL levels depend on whether the recipienthad DGF or not. To clarify the meaning of this interaction,


Table 1: Summary of baseline and clinical characteristics in kidney transplant donors and recipients (total sample and categorized by delayedor prompt graft function).

Total (𝑛 = 40) DGF (𝑛 = 18) Non-DGF (𝑛 = 22) 𝑃 valueDonor

Age (yr) 51.2 ± 11.4 51.1 ± 13.4 51.2 ± 9.9 0.172Male gender 26 (65) 14 (78) 12 (54.5) 0.125Living donor 11 (27.5) 3 (16.7) 8 (36.4) 0.165Expanded criteria donors 3 (7.5) 1 (5.6) 2 (9.1) 0.541Serum creatinine (mg/dL) 0.81 ± 0.18 0.85 ± 0.21 0.78 ± 0.16 0.318

Donor-recipientHLA mismatches 3.39 ± 1.24 3.38 ± 1.07 3.41 ± 1.46 0.941Cold ischemia time (h) 12.1 ± 7.9 15.2 ± 7.8 9.6 ± 7.3 0.035∗

Living donor 2.8 ± 0.5 2.5 ± 0.5 3.0 ± 0.5 0.204Deceased donor 16.2 ± 5.9 18.1 ± 5.1 14.1 ± 6.2 0.088

RecipientAge (yr) 49.2 ± 15.2 56.3 ± 10.9 43.3 ± 15.9 0.006∗

Male gender 26 (65) 11 (61) 15 (68) 0.641Caucasian 40 (100) 18 (100) 22 (100) —BMI (Kg/m2) 24.8 ± 4.9 26.2 ± 4.4 23.6 ± 5.0 0.091Previous transplant 2 (5) 0 (0) 2 (9.1)Time on dialysis (yr) 4.4 ± 4.7 5.6 ± 6.2 3.4 ± 2.3 0.135Pretransplant therapy

Dialysis 38 (95) 18 (100) 20 (90.9) 0.296Preemptive transplantation 2 (5) 0 (0) 2 (9.1)

Cause of kidney diseaseIgA nephropathy 7 (17.5) 2 (11.1) 5 (22.7) —Glomerulonephritis 6 (15.0) 4 (22.2) 2 (9.1) —Diabetic nephropathy 5 (12.5) 3 (16.7) 2 (9.1) —Autosomal dominant polycystic kidney disease 3 (7.5) 3 (16.7) 0 (0) —Unknown 4 (10.0) 1 (5.6) 3 (13.6) —Others 15 (37.5) 5 (27.8) 10 (45.5) —

Peak PRA (%) 5.5 ± 15.1 5.0 ± 15.0 5.9 ± 15.5 0.8530 29 (72.5) 14 (77.8) 15 (68.2) —1–25 8 (20.0) 3 (16.7) 5 (22.7) —26–75 3 (7.5) 1 (5.6) 2 (9.0) —

Current PRA (%) 2.3 ± 8.6 3.1 ± 11.7 1.6 ± 4.9 0.5850 34 (85) 15 (83.3) 19 (86.4) —1–25 5 (12.5) 2 (11.1) 3 (13.6) —26–50 1 (2.5) 1 (5.6) 0 (0) —

Induction regimenAntithymocyte globulin (ATG-F) 4 (10) 1 (5.6) 3 (13.6) 0.613Basiliximab/daclizumab 30 (75) 14 (77.8) 16 (72.7) 0.789

Immunosuppression at time of dischargeSteroids 38 (95.0) 18 (100) 20 (90.9) 0.296Tacrolimus 38 (95.0) 17 (94.4) 21 (95.5) 0.886Cyclosporine A 2 (0.05) 1 (5.6) 1 (5.6) 0.884

Values are expressed as mean ± standard deviation or absolute numbers and percentages. Comparisons between continuous variables were done usingparametric (𝑡-test) or nonparametric (Mann-Whitney) tests; associations between categorical variables were analyzed using the 𝜒2 test and Fisher’s exact testas appropriate; ∗𝑃 < 0.05.Abbreviations: HLA: human leukocyte antigen; BMI: body mass index; PRA: panel reactive antibody.


Table 2: Serial levels of serum creatinine and uNGAL through the first posttransplant week, according to graft function (delayed or prompt).

Serum Creatinine(mg/dL)Median, (IQR)

Prior transplantation 1st day∗(𝑛 = 40, 18 DGF)

2nd day(𝑛 = 40, 18 DGF)

4th day(𝑛 = 40, 18 DGF)

7th day(𝑛 = 40, 18 DGF)

DGF (𝑛 = 18) 7.5(6.0–11.7)

8.2(6.5–9.3)

7.5(5.9–8.5)

6.9(6.1–8.0)

6.4(5.3–8.9)

Non-DGF (𝑛 = 22) 7.8(5.1–9.4)

6.3(4.6–7.9)

4.3(2.8–6.1)

2.5(1.6–3.2)

1.9(1.4–2.4)

Urine NGAL (ng/mL)Median, (IQR)

3 to 6 h after surgery(𝑛 = 30, 13 DGF)

1st day∗(𝑛 = 35, 14 DGF)

2nd day(𝑛 = 35, 15 DGF)

4th day(𝑛 = 36, 15 DGF)

7th day(𝑛 = 35, 16 DGF)

DGF (𝑛 = 18) 647(328–1648)

866(500–1256)

834(510–2632)

851(549–1643)

407(106–1249)

Non-DGF (𝑛 = 22) 256(105–446)

129(64–306)

80(29–138)

47(36–91)

34(26–57)

∗1st day = 8 to 12 h after surgery; values are medians and interquartile range (25th to 75th percentile).Abbreviations: uNGAL: urinary neutrophil gelatinase associated lipocalin; IQR: interquartile range; DGF: delayed graft function; non-DGF: prompt function.

0

100

200

300

400

500

600

700

800

900

1000

Day 0 Day 1 Day 2 Day 4 Day 7

DGFNon-DGF

Med

ian

uNG

AL

leve

ls

Time point measurements (days)

Figure 1: Evolution of uNGAL levels through first week aftertransplantation, according to graft function.Abbreviations: uNGAL:urinary neutrophil gelatinase associated lipocalin; DGF: delayedgraft function.

Figure 2 shows the predicted uNGAL trajectories over timefor four hypothetical subjects: two recipients who developedDGF (one with 4 years of dialysis and one with 10 years), andtwo other patients with prompt graft function (similar timeon dialysis, 4 and 10 years). Hypothetically, the remainingvariables were equal in all four patients. The predicteduNGALvalueswere estimated using the coefficients estimatesof Table 3 (e.g., the predicted uNGAL values at the first dayfor a recipient with 4 years of dialysis with prompt function =exp [(5.46 − 2.14) + 0.94 + 0.4 + (0.076 ∗ 4 years of dialysis)]= 158 ng/mL).

0

200

400

600

800

1000

1200

1400

1600

Day 0 Day 1 Day 2 Day 4 Day 7

Estim

ated

uN

GA

L le

vels

DGF, dialysis 10 yrsDGF, dialysis 4 yrs

Non-DGF, dialysis 10 yrsNon-DGF, dialysis 4 yrs

Figure 2: Predicted uNGAL values over time of four hypotheticalsubjects, estimated from multiple linear mixed model presentedin Table 3. Abbreviations: uNGAL: urinary neutrophil gelatinaseassociated lipocalin;DGF: delayed graft function; non-DGF: promptgraft function.

3.3. Prediction of DGF by uNGAL Levels (ROC analy-sis). Receiver-operating characteristic (ROC) curves showeduNGAL on the first postoperative days were accurate inpredicting DGF (Tables 4 and 5). Table 4 displays the derivedsensitivities, specificities, and predictive values for uNGALat the cutoff concentrations that provided the maximumsum of sensitivity and specificity. Regarding the areas underthe ROC curves (AUC), the ability of uNGAL to predictDGF was moderately accurate at baseline and first day, andhighly accurate at second, fourth, and seventh days (Table 5,Figure 3). In the first two posttransplant days the diagnosticperformance of uNGAL was better than of serum creatinine,and quite similar to that of cystatin C.The reduction in serumcreatinine between first and second days resulted in AUC= 0.78 [0.64–0.92] and was worse than uNGAL for DGFprediction.


Table 3: Results of the final linear mixed model for dependent variable ln(uNGAL) (𝑛 = 171 observations derived from 40 patients).

Coefficient estimate P value 95% CIIntercept 5.46 <0.001 4.94 5.98Graft function

DGF = 0 (prompt graft function) −2.04 <0.001 −2.64 −1.44DGF = 1 (with DGF-reference) 0 — — —

TimeTime (3 to 6 h after surgery) 0.47 0.088 −0.07 1.00Time (1st day) 0.94 0.001 0.41 1.47Time (2nd day) 1.01 <0.001 0.49 1.53Time (4th day) 1.02 <0.001 0.50 1.54Time (7th day-reference) 0 — — —

Time∗DGFTime (3 to 6 h after surgery)∗DGF = 0 1.40 <0.001 0.68 2.13Time (1st day)∗ DGF = 0 0.40 0.257 −0.29 1.10Time (2nd day)∗ DGF = 0 −0.37 0.295 −1.06 0.32Time (4th day)∗DGF = 0 −0.73 0.039 −1.42 −0.03Time (7th day)∗DGF = 0 (reference) 0 — — —

Time on dialysis 0.076 0.003 0.03 0.12Abbreviations: uNGAL: urinary neutrophil gelatinase associated lipocalin; ln: natural logarithm; DGF: delayed graft function (DGF = 0, no delayed graftfunction).

Table 4: Sensitivity, specificity, and predictive values for DGF using specific uNGAL cut-off values.

Time after transplant uNGAL cutoff (ng/mL) Sensitivity (%) Specificity (%) PPV NPVShortly after surgery (3 to 6 h) 479 77 88 87 791st day (8 to 12 h after surgery) 286 100 76 81 1002nd day 277 93 90 90 934th day 232 93 95 95 937th day 63 94 84 86 93DGF: delayed graft function; uNGAL: urinary neutrophil gelatinase-associated lipocalin; PPV: positive predictive value; NPV: negative predictive value.

3.4. Independent Association of uNGAL Levels and DGF(Multivariable Analyses). Multivariable logistic regressionanalyses revealed that uNGAL levels remained independentlyassociated with DGF at most time points, after adjusting forclinically relevant risk factors for DGF (Table 6). Further-more, recipient age was the other significant independentpredictor of DGF in almost all models. To be more clinicallyrelevant, estimates of DGF risk were converted to every50 ng/mL of increase in uNGAL or per each 5 years ofincrease in age, instead of estimates per each unit of increase.

3.5. Within One-Year after Kidney Transplantation. Duringthe first year, 10 KTx recipients were rehospitalized account-ing for a total of 19 hospital admissions. There was onerehospitalization in six patients, two in two patients, threein one patient, and six rehospitalizations in one patientwith a psychological disorder and suicidal ideation. Thecauses of rehospitalization were infection in five admissions(mostly, urinary tract infection), renal dysfunction in six, andnonrenal causes in the remaining eight admissions (suicidalideation, acute pulmonary edema, and neutropenia).

Excluding the recipient with several rehospitalizationsdue to psychological decompensation, the length of hospital

stay of the remaining recipients admissions was 7 [3] days,and no significant differences were found between recipientsfrom living or deceased donors.

The acute rejection episodes were collected throughoutthe first posttransplant year. Ten recipients (25%) had anacute rejection episode during inpatient hospitalization fortransplant surgery, and only one patient was rehospitalizedone month after KTx with an acute rejection episode con-firmed by biopsy.

At one year after transplantation, all patients were alivebut two grafts were lost. At this time, the median plasmacreatinine was significantly higher in DGF group comparedto non-DGF: 1.6mg/dL [IQR: 1.2–2.5] versus 1.3mg/dL [IQR:1.0–1.5], 𝑃 = 0.049.

3.6. Prognostic Value of First-Week uNGAL Levels in One-YearGraft Function. Thecorrelation between uNGAL collected inthe first week after KTx and serum creatinine at one yearwas explored. Except for uNGAL collected within the first24 h after transplantation, uNGAL levels were positively cor-related with serum creatinine evaluated at time of discharge,and also at 1, 3, 6, and 12 months. Likewise, uNGAL levels at


Table 5: Area under the receiver-operating characteristic curve ateach time point for uNGAL, serum creatinine, and serum cystatinC for predicting DGF.

Time aftertransplant AUC (95% CI) 𝑃 value

Urine NGAL(ng/mL)

Shortly aftersurgery (3 to 6 h) 0.77 (0.58–0.97) 0.010

1st day (8 to 12 hafter surgery) 0.88 (0.77–1.0) <0.001

2nd day 0.96 (0.90–1.0) <0.0014th day 0.99 (0.97–1.0) <0.0017th day 0.93 (0.86–1.0) <0.001

Serumcreatinine(mg/dL)

Priortransplantation 0.56 (0.38–0.74) 0.514

1st day (8 to 12 hafter surgery) 0.77 (0.61–0.93) 0.007


Serumcystatin C(mg/L)

1st day (6 to 12 hafter surgery) 0.90 (0.79–1.0) <0.001


DGF: delayed graft function; uNGAL: urinary neutrophil gelatinase-associated lipocalin.

days 2, 4, and 7 were inversely correlated with eGFR at 6 and12months (data not shown).

Theprognostic value of early uNGALvalues on long-termallograft function (one year after KTx) was tested by mul-tivariable analysis. In multivariable linear regression modelsfor serum creatinine at 12months, uNGAL measured on thefourth and seventh days were independently associated withone-year graft function, adjusting for established variablesthat usually affect graft function, including acute rejectionepisodes and rehospitalizations that occurred during the firstposttransplant year (Table 7).

4. Discussion

Themajor finding of this study is that uNGAL is a promisingbiomarker for allograft dysfunction that can be easily andnoninvasively assayed in the early posttransplant period. Weprospectively evaluated uNGAL in a cohort of 40 kidneyallograft recipients during the first posttransplant week. At allmeasured timepoints, uNGAL levelswere consistently higherin patients who developed DGF, including the earliest levelsobtained from the first urine sample collected approximately3 to 6 h after transplant surgery. At this time, clinical diagnosisof DGF is yet not possible, but a simple and noninvasivetest can already recognize kidney dysfunction and stratifypatients according to likelihood of requiring posttransplantdialysis.

It would be ideal to diagnose graft dysfunction with anearly and highly sensitive biologic marker of renal tubular

1.00.80.60.40.20.0

Sens

itivi

ty

1.0

0.8

0.6

0.4

0.2

0.0

Reference line

Creat2

uNGAL2CystC1Creat1uNGAL1

ROC curve

Source of the curve

CystC 2

Creat2 Creat1

False positive rate (1 − specificity)

Figure 3: Receiver-operating characteristic curves for uNGAL,serum creatinine and changes in serum creatinine, and serumcystatin C measured at posttransplant days 1 and 2 for predictingdelayed graft function. Abbreviations: uNGAL: urinary neutrophilgelatinase associated lipocalin; Creat: serum creatinine; Cyst: serumCystatinC;Creat2-Creat1: creatinine reduction rate between the firstand the second day.

injury. One of the most promising markers is NGAL, and ourfindings provide further information for the use of uNGAL asa diagnostic and prognostic tool for DGF. According to ourestimation, uNGAL values shortly after transplant surgerywill be much higher in patients who went on to developDGF and will rise further in the following days. In contrast,patients with prompt function will have lower levels, whichdecrease consistently along the week.The kinetics of changesin these recipients compared to those who presented DGF isquite different. It seems that, not only the baseline levels, butalso the pattern of uNGAL longitudinal changes can reflectgraft dysfunction.

The association between higher NGAL levels and DGFafter KTx has been previously published [16, 20, 22, 24, 25].But the findings are not consistent regarding the kinetics ofuNGAL according to DGF. Hollmen et al. [22] found initiallevels of uNGAL higher inDGF patients, but on the followingday a decrease was observed, as it happened with recipientswith prompt function. As mentioned before, our studydid not confirm this declining in DGF patients. Recipientswho went on to develop DGF had initial higher levels ofuNGAL that rise further on the following posttransplantdays, differing from patients with prompt graft function.Our findings are in agreement with results reported by Hall


Table 6: Association of uNGAL with delayed graft function by multivariable analysis (logistic regression).

Delayed graft functionOR adjusted∗ 𝑃 value 95% CI

Model 1 (uNGAL at 3 to 8 h after surgery)uNGAL0 (per 50 ng/mL of increase) 1.15 0.044 1.01–1.31Recipient age (per 5 years of increase) 1.49 0.054 0.99–2.24

Model 2 with (uNGAL at day 1)uNGAL1 (per each 50 ng/mL of increase) 1.22 0.012 1.05–1.42Recipient age (per 5 years of increase) 1.99 0.022 1.11–3.57

Model 3 (uNGAL at day 2)uNGAL2 (per each 50 ng/mL of increase) 1.35 0.004 1.10–1.66

Model 4 (uNGAL at day 4)uNGAL4 (1 ng/mL increase) 3.01 0.035 1.08–8.40

Model 5 (uNGAL at day 7)uNGAL7 (per each 50 ng/mL of increase) 1.43 0.050 1.01–2.04Recipient age (per 5 years of increase) 1.73 0.038 1.03–2.90

Note: results given by logistic regression (backward Wald test).Abbreviations: OR: odds ratio; 95% CI (95% confidence interval); uNGAL: urinary neutrophil gelatinase-associated lipocalin.∗Adjusted for pretransplant time on dialysis, recipient gender and age, and donor age.

Table 7: Significant factors associated with serum creatinine at one year after kidney transplantation.

Regression coefficient adjusted∗ 𝑃-value 95% CIModel with uNGAL at day 4

Donor gender (male versus female) 0.042 0.004 0.015–0.069Donor age (years) 0.011 0.008 0.003–0.020uNGAL4 (ln, ng/mL) 0.067 0.045 0.002–0.132

Model with uNGAL at day 7Time on dialysis (ln, ng/mL) 0.042 0.004 0.015–0.069Donor age (years) 0.018 0.002 0.008–0.029uNGAL7 (ln, ng/mL) 0.138 0.007 0.041–0.235

Note: results given by multiple linear regression; serum creatinine (ln) at 12 months as the dependent variable. Only the significant variables associated withserum creatinine are displayed. Abbreviations: uNGAL: urinary neutrophil gelatinase-associated lipocalin.∗Adjusted for donor status, donor age, recipient age and gender, pretransplant time on dialysis, rehospitalizations, and acute rejection episodes throughout thefirst year.

et al. [24]. It seems that, above and beyond the markedlyhigher levels of uNGAL in patients with graft dysfunction,the contrasting pattern of uNGAL longitudinal changes candistinguish recipients who will need dialysis in the first weekposttransplantation.

To the best of our knowledge, this is the first reportthat used linear mixed analysis in describing longitudinalchanges of uNGAL in the first week following KTx. Multipleobservations of a variable on a particular patient are likelyto be positively correlated, so they should not be treated asindependent measurements. Although models that take thisdesign into consideration are more complicated, they are alsomore specific and powerful since they permit the study ofchanges over time. Linear mixed analysis not only permitsto model individual changes over time, but also is able todistinguish within-subject from between-subject sources ofvariation [31].

In accordance with previously published data [16, 20–22, 24], we confirmed the good performance of NGAL in pre-dicting graft dysfunction in the early posttransplant period.

Using ROC analysis, our study also corroborates uNGALas a good diagnostic marker on identifying patients withgraft dysfunction and who subsequently required dialysis.TheAUC-ROC for uNGALwasmoderately accurate forDGFprediction within the first day after transplant, and it wasexcellent at day 2 and day 4. We also determined the pairedsensitivity and specificity for the cutoff value of uNGAL,calculated to be closest to the left upper corner of the ROCspace to predict DGF. At 8 to 12 h after surgery, a cutoff of286 ng/mL had 100% sensitivity and 76% specificity for theidentification of DGF. Within the second day, uNGAL levelshigher than 277 ng/mL predicted DGF with a sensitivityof 93% and specificity of 90%. Other studies showed alsoimpressive results. Parikh et al. [16] in a study that included 53patients undergoing KTx, measured NGAL in urine samplescollected within the first 24 h following transplantation andreported an AUC-ROC of 0.9, similar to ours obtained 8to 12 h after surgery. Another study [24] conducted in 91recipients evaluated uNGAL within 6 h after transplantationand predicted subsequent DGF with an AUC-ROC = 0.81.


Most recently, Hollmen et al. [22] undertook a large cohortstudy that included 176 KTx recipients. Urine was collectedbefore transplant, at then at days 1, 3, 7, and 14, and uNGALwas measured at each time point. The authors found andAUC-ROC = 0.74 at day 1.

We report a superior performance of uNGAL level forpredicting DGF over serum creatinine measured at the sametime. UrinaryNGALmeasured at the first day predictedDGFwith an AUC-ROC of 0.93, which is markedly better than anAUC-ROC = 0.76 shown by serum creatinine measured inthe same day, and also than an AUC-ROC = 0.83 obtainedfrom creatinine reduction ratio from first to second day,but quite similar to cystatin C (0.95), a marker consideredmore accurately to detect changes in renal function [32–35]. Furthermore, our analyses also revealed that uNGALlevels predicted DGF, even after adjusting for pretransplantvariables known to be traditionally associated with DGF.

Besides DGF, the other factors that significantly influ-enced uNGAL levels were previous time on dialysis, recip-ient’s age at time of transplantation and cold ischemiatime. These three variables were positively correlated withuNGAL values. Mishra and coworkers [17] have shownthat the immunohistochemical staining intensity for NGALwas strongly correlated with cold ischemia time and NGALexpression was significantly increased in deceased donorbiopsies. We found that uNGAL levels were higher in graftrecipients fromdeceased donors, but only significantly higherat the second day. It is known that prolonged cold preserva-tion of kidneys can lead to severe injury, which is critical inthe success of deceased-donor kidney transplantation [36].However, there is a progressive effort of our transplant teamto avoid prolonged cold preservation. Maybe this attemptattenuated the effect of cold ischemic injury in kidneysfrom deceased donors, which become comparable to livingdonors concerning uNGAL values. An interesting findingof our study was that uNGAL levels at all-time points werecorrelated with length of hospital stay. It is well knownthat the occurrence of DGF prolongs the recipient’s hospitalstay. And it is worthy of note to realize that patients withearly higher levels of uNGAL will expect longer time ofhospitalization, probably due to graft dysfunction.

As other studies [16–19, 24, 26, 37], we confirmed thatuNGAL levels were inversely correlated with eGFR andpositively correlated with serum creatinine at each measuredtime point. We also showed that not only in the first week,but longer after that, uNGAL levels measured in the firstseven days after KTx were still predictive of graft functionthroughout the first year after transplantation. Even afteradjusting for donor status, acute rejection episodes, hos-pitalizations occurred in the first year, and other knownvariables that usually affect graft function, uNGAL evaluatedat days four and seven were predictive of one-year serumcreatinine, which can be considered a surrogate marker oflong-term graft survival [38, 39]. In contrast, Hollmen andcoworkers study [22] did not find any correlation betweenuNGAL and renal function at one year. In their study, uNGALcollected in the first two weeks after transplantation was onlycorrelated with renal function up to 3months. Our results donot corroborate this lack of correlation and are in agreement

with a recent study that also associated perioperative uNGALlevels to one-year allograft function [28].

Our study has several strengths. First, it is a prospective-cohort design study. Second, we measured uNGAL at severaltime points within the first posttransplant week, and not atone single point. Longitudinal studies have the advantage ofproviding detailed information about how a marker changesover time; however the studies present some statistical com-plexities, since the customary assumption that all observa-tions are independent usually does not hold. And this was thethird strength of our study: the use of longitudinal methodsto handle the serial changes of uNGAL. A fourth strength wasthe technical determination that we have chosen to measureNGAL. We used a commercially available kit for uNGALdetermination (Abbott Architect NGAL), which is simple toimplement in routine practice and it is considered one of thebest methods for detecting acute kidney injury [40].

Similarly to other authors [16, 22], we have chosen tomeasure NGAL in urine, instead of blood, since uNGALrepresents tubule damage in the kidney rather than filtrationfrom blood [14, 41]. An increased level of NGAL in urineusually indicates injury of proximal tubular cells and seemsto be more specific compared to serum NGAL, which can beproduced by other organs and released into the circulationfollowing a transplant surgery [42]. Other advantages ofurinary diagnostics include the noninvasive nature of samplecollection and the reduced number of interfering protein[43]. However, despite the undoubtedly value of urinarymarkers of kidney injury, their use in transplant recipientscan be also a drawback because of possible transient graftanuria, which may preclude the availability of urine andconsequently the lack of sample to measure NGAL. Dueto the shortcoming of urine biomarkers in anuric KTxrecipients, some studies have evaluated the performance ofserum/plasma NGAL in predicting graft function recoveryafter KTx [21, 27]. As we did not measure serum/plasmaNGAL values, we could not compare their effectiveness inpredicting DGF in our sample. In our study, 4 or 5 recipientswere anuric in each measurement, resulting in 12% of ourpatients not having urine sample to determine uNGAL in thatparticular time point. The measurement of serum/plasmaNGAL could have been a valuable alternative in theserecipients, since it could also be obtained noninvasively inpatients who required dialysis during the transient period ofanuria.

Similar to other areas in medicine, in kidney transplan-tation early diagnosis and timely intervention will improveoutcomes. Ischemic injury of the renal allograft is a criticalearly insult that increases the risk of acute tubular necrosisand long-term graft loss. If DGF could be detected inthe early hours after surgical procedure, maybe a tailoredand more individualized intervention could be achieved.Perioperative fluid management must ensure the restorationand maintenance of the intravascular volume, in order toobtain an appropriate graft function. Aggressive hydrationhas been recognized to be effective in avoiding DGF, butfluid overload may also precipice the need of dialysis withthe risk of hipovolemia and consequent renal ischemia.Early identification of DGF patients could allow to be more


judicious and to modify postoperative fluid managementin favor of maintaining just adequate filling pressures tomaintain adequate intravascular volume and prevent fluidoverload [44]. Regarding immunosuppression, the inductionprotocol chosen for this group of patients should have theassociated effect of decreasing DGF rates, by suppressingleukocyte-rich vascular congestion and endothelial injury,and the introduction of calcineurin inhibitors could beavoided or delayed due to their vasoconstrictive properties.Cytomegalovirus (CMV) infection has also direct and indi-rect effects on transplant graft function, and some previousevidence has been published relating the association betweenthe use of ganciclovir and the lower occurrence of DGF[45]. Nowadays, the prophylaxis with valganciclovir shouldbe other aspect taken into account in recipients that we knowthey will develop DGF, since this prophylaxis may do morethan just delay the occurrence of CMV disease.

Several studies were done in renal transplantation toidentify early biomarkers for the diagnosis of DGF. However,there is still no routine application of any of these markersin clinical transplantation. The present study clearly supportthat uNGAL represents an early marker of graft injury andis strongly associated with dialysis-based diagnosis of DGFand one-year graft function. Other studies are necessary toclarify the genesis and sources of plasma and urinary NGALand validate the accuracy of uNGAL as a diagnosticmarker ofrenal graft injury and predictor for DGF in assorted centres,across different practices and sets of variables.

Conflict of Interests

The authors declare that there is no conflict regarding thepublication of this paper. The results presented in this paperhave not been published previously in whole or part, neitherin abstract format.

Authors’ Contribution

Isabel Fonseca designed the study, collected data, compiledthe database, performed statistical analyses, interpreted thedata, and wrote the paper. Jose Carlos Oliveira andMadalenaCruz were responsible for the assays. Manuela Almeida,Anabela Malho, La Salete Martins, Sofia Pedroso, LeonıdioDias, Josefina Santos, and Antonio Castro Henriques wereinvolved in patients recruiting patients and coordinated thecollection of samples and assembling the patient information.Antonio Castro Henriques and Denisa Mendonca helped todesign and coordinate the study. Jose Carlos Oliveira, LuısaLobato, Antonio Castro Henriques, and Denisa Mendoncarevised the drafts and gave helpful comments for analysisand interpretation of the data. All authors approved the finalversion.

Acknowledgments

The authors recognize and thank Abbott Laboratories fortheir valuable contribution for donating kits used for test-ing almost 200 samples. The remaining kits were financed

by funds of Unit for Multidisciplinary Investigation inBiomedicine, Porto, Portugal. The investigators also thankDr. Antonio Cabrita for allowing the study, and especially tonursing and laboratory staff for their commitment and crucialhelp on samples collection.

References

[1] N. Perico, D. Cattaneo,M.H. Sayegh, andG. Remuzzi, “Delayedgraft function in kidney transplantation,” Lancet, vol. 364, no.9447, pp. 1814–1827, 2004.

[2] A.O.Ojo, R.A.Wolfe, P. J.Held, F. K. Port, andR. L. Schmouder,“Delayed graft function: risk factors and implications for renalallograft survival,” Transplantation, vol. 63, no. 7, pp. 968–974,1997.

[3] O. H. J. Koning, J. H. van Bockel, F. J. van der Woude, G. G.Persijn, J. Hermans, and R. J. Ploeg, “Risk factors for delayedgraft function in University of Wisconsin solution preservedkidneys from multiorgan donors,” Transplantation Proceedings,vol. 27, no. 1, pp. 752–753, 1995.

[4] M. T. Sellers, M. H. Gallichio, S. L. Hudson et al., “Improvedoutcomes in cadaveric renal allografts with pulsatile preserva-tion,” Clinical Transplantation, vol. 14, no. 6, pp. 543–549, 2000.

[5] D. W. Gjertson, “Impact of delayed graft function and acuterejection onkidney graft survival,”Clinical Transplants, pp. 467–480, 2000.

[6] S. G. Yarlagadda, S. G. Coca, A. X. Garg et al., “Marked variationin the definition and diagnosis of delayed graft function: asystematic review,”Nephrology Dialysis Transplantation, vol. 23,no. 9, pp. 2995–3003, 2008.

[7] H. Boom,M. J. K.Mallat, J.W.De Fijter, A.H. Zwinderman, andL.C. Paul, “Delayed graft function influences renal function, butnot survival,” Kidney International, vol. 58, no. 2, pp. 859–866,2000.

[8] M. Giral-Classe, M. Hourmant, D. Cantarovich et al., “Delayedgraft function of more than six days strongly decreases long-term survival of transplanted kidneys,” Kidney International,vol. 54, no. 3, pp. 972–978, 1998.

[9] C. Y. Lu, J. G. Penfield, M. L. Kielar, M. A. Vazquez, and D. R.Jeyarajah, “Hypothesis: is renal allograft rejection initiated bythe response to injury sustained during the transplant process,”Kidney International, vol. 55, no. 6, pp. 2157–2168, 1999.

[10] D. A. Shoskes and J. M. Cecka, “Deleterious effects of delayedgraft function in cadaveric renal transplant recipients indepen-dent of acute rejection,” Transplantation, vol. 66, no. 12, pp.1697–1701, 1998.

[11] C. Troppmann, K. J. Gillingham, R. W. G. Gruessner et al.,“Delayed graft function in the absence of rejection has no long-term impact. A study of cadaver kidney recipients with goodfunction at 1 year after transplantation,” Transplantation, vol. 61,no. 9, pp. 1331–1337, 1996.

[12] R. L. Mehta and G. M. Chertow, “Acute renal failure definitionsand classification: time for change?” Journal of the AmericanSociety of Nephrology, vol. 14, no. 8, pp. 2178–2187, 2003.

[13] D. Bolignano, G. Coppolino, A. Lacquaniti, and M. Buemi,“From kidney to cardiovascular diseases: NGAL as a biomarkerbeyond the confines of nephrology,”European Journal of ClinicalInvestigation, vol. 40, no. 3, pp. 273–276, 2010.

[14] K. Mori, H. T. Lee, D. Rapoport et al., “Endocytic delivery oflipocalin-siderophore-iron complex rescues the kidney from


ischemia-reperfusion injury,” Journal of Clinical Investigation,vol. 115, no. 3, pp. 610–621, 2005.

[15] J. Mishra, M. A. Qing, A. Prada et al., “Identification of neu-trophil gelatinase-associated lipocalin as a novel early urinarybiomarker for ischemic renal injury,” Journal of the AmericanSociety of Nephrology, vol. 14, no. 10, pp. 2534–2543, 2003.

[16] C. R. Parikh, A. Jani, J. Mishra et al., “Urine NGAL and IL-18are predictive biomarkers for delayed graft function followingkidney transplantation,” American Journal of Transplantation,vol. 6, no. 7, pp. 1639–1645, 2006.

[17] J. Mishra, Q. Ma, C. Kelly et al., “Kidney NGAL is a novel earlymarker of acute injury following transplantation,” PediatricNephrology, vol. 21, no. 6, pp. 856–863, 2006.

[18] J. Malyszko, J. S. Malyszko, and M. Mysliwiec, “Serum neu-trophil gelatinase-associated lipocalin correlates with kidneyfunction in renal allograft recipients,” Clinical Transplantation,vol. 23, no. 5, pp. 681–686, 2009.

[19] J. Malyszko, J. S. Malyszko, H. Bachorzewska-Gajewska, B.Poniatowski, S. Dobrzycki, and M. Mysliwiec, “Neutrophilgelatinase-associated lipocalin is a new and sensitive marker ofkidney function in chronic kidney disease patients and renalallograft recipients,” Transplantation Proceedings, vol. 41, no. 1,pp. 158–161, 2009.

[20] U. Lebkowska, J. Malyszko, A. Lebkowska et al., “Neutrophilgelatinase-associated lipocalin and cystatin C could predictrenal outcome in patients undergoing kidney allograft trans-plantation: a prospective study,” Transplantation Proceedings,vol. 41, no. 1, pp. 154–157, 2009.

[21] M. Kusaka, Y. Kuroyanagi, T. Mori et al., “Serum neutrophilgelatinase-associated lipocalin as a predictor of organ recoveryfrom delayed graft function after kidney transplantation fromdonors after cardiac death,” Cell Transplantation, vol. 17, no. 1-2,pp. 129–134, 2008.

[22] M. E. Hollmen, L. E. Kyllonen, K. A. Inkinen,M. L. T. Lalla, andK. T. Salmela, “Urine neutrophil gelatinase-associated lipocalinis a marker of graft recovery after kidney transplantation,”Kidney International, vol. 79, no. 1, pp. 89–98, 2011.

[23] M. E. Hollmen, L. E. Kyllonen, K. A. Inkinen, M. L. T. Lalla,J. Merenmies, and K. T. Salmela, “Deceased donor neutrophilgelatinase-associated lipocalin and delayed graft function afterkidney transplantation: a prospective study,” Critical Care, vol.15, no. 3, article R121, 2011.

[24] I. E. Hall, S. G. Yarlagadda, S. G. Coca et al., “IL-18 andurinary NGAL predict dialysis and graft recovery after kidneytransplantation,” Journal of the American Society of Nephrology,vol. 21, no. 1, pp. 189–197, 2010.

[25] A. Bataille, S. Abbas, O. Semoun et al., “Plasma neutrophilgelatinase-associated lipocalin in kidney transplantation andearly renal function prediction,” Transplantation, vol. 92, no. 9,pp. 1024–1030, 2011.

[26] N. E. Magnusson, M. Hornum, K. A. Jorgensen et al., “Plasmaneutrophil gelatinase associated lipocalin (NGAL) is associatedwith kidney function in uraemic patients before and afterkidney transplantation,”BMCNephrology, vol. 13, article 8, 2012.

[27] M. Kusaka, F. Iwamatsu, Y. Kuroyanagi et al., “Serumneutrophilgelatinase associated lipocalin during the early postoperativeperiod predicts the recovery of graft function after kidneytransplantation from donors after cardiac death,” Journal ofUrology, vol. 187, no. 6, pp. 2261–2267, 2012.

[28] I. E. Hall, M. D. Doshi, P. P. Reese, R. J. Marcus, H. Thiessen-Philbrook, and C. R. Parikh, “Association between peritrans-plant kidney injury biomarkers and 1-year allograft outcomes,”

Clinical Journal of the American Society of Nephrology, vol. 7, no.8, pp. 1224–1233, 2012.

[29] A. D. Rule, T. S. Larson, E. J. Bergstralh, J. M. Slezak, S. J.Jacobsen, and F. G. Cosio, “Using serum creatinine to estimateglomerular filtration rate: accuracy in good health and inchronic kidney disease,” Annals of Internal Medicine, vol. 141,no. 12, pp. 929–937, 2004.

[30] U. Poge, T. Gerhardt, B. Stoffel-Wagner et al., “Can modi-fications of the MDRD formula improve the estimation ofglomerular filtration rate in renal allograft recipients?”Nephrol-ogy Dialysis Transplantation, vol. 22, no. 12, pp. 3610–3615, 2007.

[31] G. M. Fitzmaurice and C. Ravichandran, “A primer in longitu-dinal data analysis,” Circulation, vol. 118, no. 19, pp. 2005–2010,2008.

[32] E. Premaratne, R. J. MacIsaac, S. Finch, S. Panagiotopoulos,E. Ekinci, and G. Jerums, “Serial measurements of cystatin care more accurate than creatinine-based methods in detectingdeclining renal function in type 1 diabetes,” Diabetes Care, vol.31, no. 5, pp. 971–973, 2008.

[33] L. A. Stevens, J. Coresh, C. H. Schmid et al., “Estimating GFRusing serum cystatin C alone and in Combination with serumcreatinine: a pooled analysis of 3,418 individuals with CKD,”American Journal of Kidney Diseases, vol. 51, no. 3, pp. 395–406,2008.

[34] B. A. Perkins, R. G. Nelson, B. E. P. Ostrander et al., “Detectionof renal function decline in patients with diabetes and normalor elevated GFR by serial measurements of serum cystatin Cconcentration: results of a 4-year follow-up study,” Journal ofthe American Society of Nephrology, vol. 16, no. 5, pp. 1404–1412,2005.

[35] O. F. Laterza, C. P. Price, and M. G. Scott, “Cystatin C:an improved estimator of glomerular filtration rate?” ClinicalChemistry, vol. 48, no. 5, pp. 699–707, 2002.

[36] A. K. Salahudeen andW.May, “Reduction in cold ischemia timeof renal allografts in the United States over the last decade,”Transplantation Proceedings, vol. 40, no. 5, pp. 1285–1289, 2008.

[37] M. Szewczyk, T. Wielkoszynski, M. Zakliczynski, and M.Zembala, “Plasma neutrophil gelatinase-associated lipocalin(NGAL) correlations with cystatin C, serum creatinine, andglomerular filtration rate in patients after heart and lungtransplantation,” Transplantation Proceedings, vol. 41, no. 8, pp.3242–3243, 2009.

[38] I. Fonseca, M. Almeida, L. S. Martins et al., “First-year renalfunction predicts long-term renal allograft loss,” Transplanta-tion Proceedings, vol. 43, no. 1, pp. 106–112, 2011.

[39] S. Hariharan, M. A. McBride,W. S. Cherikh, C. B. Tolleris, B. A.Bresnahan, and C. P. Johnson, “Post-transplant renal functionin the first year predicts long-term kidney transplant survival,”Kidney International, vol. 62, no. 1, pp. 311–318, 2002.

[40] E. Cavalier, A. C. Bekaert, A. Carlisi, D. Legrand, J. M.Krzesinski, and P. Delanaye, “Neutrophil gelatinase-associatedlipocalin (NGAL) determined in urinewith theAbbott architector in plasma with the Biosite Triage? The laboratory’s point ofview,” Clinical Chemistry and Laboratory Medicine, vol. 49, no.2, pp. 339–341, 2011.

[41] T. Kuwabara, K. Mori, M. Mukoyama et al., “Urinary neu-trophil gelatinase-associated lipocalin levels reflect damageto glomeruli, proximal tubules, and distal nephrons,” KidneyInternational, vol. 75, no. 3, pp. 285–294, 2009.

[42] J. B. Cowland, O. E. Sørensen, M. Sehested, and N. Borregaard,“Neutrophil gelatinase-associated lipocalin is up-regulated in


human epithelial cells by IL-1 beta, but not by TNF-𝛼,” Journalof Immunology, vol. 171, no. 12, pp. 6630–6639, 2003.

[43] P. Devarajan, “Neutrophil gelatinase-associated lipocalin: apromising biomarker for human acute kidney injury,” Biomark-ers in Medicine, vol. 4, no. 2, pp. 265–280, 2010.

[44] P. Schnuelle and F. Johannes van der Woude, “Perioperativefluidmanagement in renal transplantation: a narrative review ofthe literature,” Transplant International, vol. 19, no. 12, pp. 947–959, 2006.

[45] J. A. Fishman, V. Emery, R. Freeman et al., “Cytomegalovirusin transplantation—challenging the status quo,” Clinical Trans-plantation, vol. 21, no. 2, pp. 149–158, 2007.

Submit your manuscripts athttp://www.hindawi.com

Stem CellsInternational

Hindawi Publishing Corporationhttp://www.hindawi.com Volume 2014


MEDIATORSINFLAMMATION

of


Behavioural Neurology

EndocrinologyInternational Journal of



Disease Markers


BioMed Research International

OncologyJournal of



Oxidative Medicine and Cellular Longevity


PPAR Research

The Scientific World JournalHindawi Publishing Corporation http://www.hindawi.com Volume 2014

Immunology ResearchHindawi Publishing Corporationhttp://www.hindawi.com Volume 2014

Journal of

ObesityJournal of



Computational and Mathematical Methods in Medicine

OphthalmologyJournal of


Diabetes ResearchJournal of



Research and TreatmentAIDS


Gastroenterology Research and Practice


Parkinson’s Disease

Evidence-Based Complementary and Alternative Medicine

Volume 2014Hindawi Publishing Corporationhttp://www.hindawi.com

research article neutrophil gelatinase-associated lipocalin in...

Documents